Algoritam temeljen na promjeni orijentacije za sustav za detekciju pada starijih osoba

master's thesis

• Title (english): Orientation based algorithm for fall detection system for the elderly
• Author: Andrej Žele
• Mentor: Igor Lacković (mentor)
• Committee member: Igor Lacković (predsjednik povjerenstva)
• Committee member: Ratko Magjarević (član povjerenstva)
• Committee member: Hrvoje Džapo (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-09-27, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Ljudi su danas u svakodnevnom životu u opasnosti od raznih oblika ozljeda. Istraživanja pokazuju da svaka treća osoba starija od 65 godina barem jednom u godini padne.
Uređaj koji se koristio za detekciju pada prilikom izrade ovog rada je Shimmer3. Mogućnost detektiranja pada pomoću promjene orijentacije vektora akceleracije nudi dobar i inventivan način mjerenja pokretljivosti. Algoritam koji je ponuđen zasnovan je na dva parametra: gravitacijom otežanu amplitudu akceleracija te promjenu kuta orijentacije vektora akceleracije. Izvršena su mjerenja uobičajenih aktivnosti i simuliranih padova na 17 ispitanika uz prethodno pripremljen mjerni sustav. Provedena je analiza ovog algoritma uz ispitivanje pragova pomoću ROC krivulja.
Ostavljena je mogućnost unapređenja algoritma uz ispitivanje dodatnih parametara.

Abstract (english)

People today in everydaylife are always exposed to dangers and risks of many different injuries. Researches show that every third person older than 65 years falls down at least once per year.
Device that was used throughout this assignment is the Shimmer3. The possibility of detecting fall using the switch of the orientation of the acceleration vector was something that offered a good and inventive way of measuring mobility. The fall detection algorithm was based on two parameters: gravity-weighted sum vector magnitude and the tilted angle between the accelerometer y-axis and the vertical direction. We performed multiple mobility measurments on 17 examinees. An analysis of this algorithm was carried out with the examination of the thresholds using the ROC curve.
There are also possibilities of upgrading the algorythm using more paramaters.

• Keywords: stari ljudi; Shimmer; detekcija pada; Matlab
• Keywords (english): old people; Shimmer; fall detection
• Language: croatian
• URN:NBN: urn:nbn:hr:168:042165
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-21 14:41:10